Synopsis On 09 November 1999, the loaded bulk carrier Alcor was upbound for Trois-Rivires, Quebec, on the St. Lawrence River, under the conduct of a pilot. At 1444 local time, while undertaking a course alteration to starboard, the vessel ran aground near the eastern end of le d'Orlans. A refloating attempt the next evening succeeded in freeing the vessel, but only briefly, and the vessel grounded a second time near the initial grounding position. The Alcorsustained major hull damage near midships due to bending forces incurred during successive low-tide cycles. The damaged hull was temporarily repaired and roughly half of the cargo was discharged onto smaller vessels. On December5, the Alcorwas refloated and conducted to the port of Qubec. It was declared a constructive total loss. While the Alcor was being refloated and later, while the vessel was upbound with the assistance of tugs, the Traverse du Nord section of the river was temporarily closed. The closure caused several downbound vessels to anchor upriver. The subsequent re-opening of the channel resulted in a confluence of vessels wishing to depart their anchorages. During this time, a near-collision occurred between the tanker Eternity,under way, and the container ship CanmarPride, at anchor. Ce rapport est galement disponible en franais. A 1.0 Factual Information A 1.1 Particulars of the Vessel A 1.1.1 Description of the Vessel The Alcorwas a single-deck, dry bulk cargo vessel of all-welded steel construction. The propulsion machinery, electro-hydraulic steering gear, wheelhouse and crew accommodation were all arranged at the after end of the vessel. It was manoeuvred by a single balanced centre-line rudder. The location of the five cargo holds, water ballast tanks and oil fuel tanks is shown in AppendixC (OutlineGeneralArrangement). Cargo hold3 could also be used as a water ballast (deep) tank. The hull was subdivided by seven transverse watertight bulkheads, together with an inner bottom watertight tank top, which extended fore and aft throughout the cargo holds and the engine room. The main deck, upper wing tanks, inner bottom and bottom shell plating were framed longitudinally in way of the cargo holds. The single side shell plating in way of the cargo holds and upper wing ballast tanks was framed transversely. A 1.2 History of the Initial Grounding A 1.2.1 The Transit On 30 October 1999, the Alcordeparted Venezuela, bound for Trois-Rivires, Quebec. The voyage north was made through heavy weather, with the vessel pitching and rolling heavily. Deck log book entries show winds of force 5to7(17to33knots), with two to three metre swells. On 05November1999, the vessel reduced speed for a time to reduce rolling/pitching and wave impact forces. By 09 November 1999, the vessel had reached the Les Escoumins pilot boarding station in the St. Lawrence River. Draughts at this time were reported as 10.02m forward and 9.95m aft. At 05153, a river pilot boarded to take conduct of the vessel up to Qubec, whereupon, after a scheduled pilot change, the vessel was to be conducted to Trois-Rivires. The voyage upriver was uneventful, with the vessel proceeding at full manoeuvring speed. In deep water, without the effects of wind or tide, this would have given a speed of approximately 11knots. Initially, the ebb tide slowed the vessel's speed over the bottom, at times to 7or8knots.4 As the vessel progressed upriver, the flood tide gradually caught up with the vessel and her speed over the bottom increased. By 1400, the vessel had reached Sault-au-Cochon and her speed was about 13knots. At approximately 1415, the Alcorentered the more restricted waters of the North Channel (traversedunord) near Cap Gribane. As the vessel settled on the leading lights of 213True(T), the pilot confirmed his belief that the steering gyro compass repeater was showing 1low. At the next course alteration point, CapBrl, to the leading lights of 204T, the pilot ordered a course of 201gyro(G) to compensate for the 1low steering repeater and the flood tide acting in a southwesterly direction. Approaching buoy K-108, the vessel was approximately 60m to starboard of the centre of the channel, and now making 14 knots. The current was setting approximately 215T at between 2.5and 3.5knots. Buoy K-108 was the location of the next course alteration, to starboard, to make the next set of leading lights, also of 213T. The only other river traffic in the vicinity was a downbound vessel ahead at between 1and 2nautical miles (nm). (SeeFigure4 for General Area Chart.) A 1.2.2 The Grounding The bridge watch at this time consisted of the second officer, who was the officer of the watch (OOW), the pilot, who had conduct of the vessel, and the helmsman, who was at the main steering console carrying out helm orders. The master was on the port bridge wing and the door separating the bridge wing exterior and the wheelhouse interior was ajar. Between 1437 and 1438, with the vessel's bow about one cable below buoy K-108, the pilot requested a course alteration. According to some accounts, this was a course-to-steer order of 212G, while according to other accounts, it was a helm order starboard ten. The helm was placed 10 to 15 to starboard by various accounts. Soon after the helm was put over to starboard, the pilot requested more helm by using the command more. The helmsman looked to the OOW for guidance and the OOW directed him to increase helm by 5. He immediately put 5 more helm to starboard. The pilot, seeing the ship's head remain immobile, requested more helm by repeating more. The helmsman applied another 5 of starboard helm. Some accounts of the events have the pilot requesting, for a third time, more - and an additional 5 helm was applied at this time. All concur, however, that the vessel's head was now starting to swing very slowly to starboard and the vessel's rudder was between 20 and 30 to starboard. Unsatisfied with the vessel's rate of turn, the pilot removed the helmsman from his position at the steering console and took the wheel himself. According to one account, the pilot then turned the wheel to starboard once or twice to bring the rudder hard to starboard, 35. However, according to other accounts, the pilot turned the wheel to starboard five to seven times, in a highly energetic fashion. Either action would have produced the same result, 35 of starboard helm. (The wheel had a slip clutch, whereby any excess turn of the wheel past the maximum did not impart further electrical signals to the steering gear.) Having turned the wheel to starboard, the pilot immediately started turning the wheel quickly in the opposite direction and put the wheel 35 to port. The pilot noticed that the rudder angle indicator remained at 35 to starboard, and he informed the OOW of this. Some accounts have the OOW and the pilot both trying the non-follow-up (NFU) lever at this point, but others do not. Leaving the steering console, the pilot went to the very high frequency (VHF) radio to warn the downbound vessel of their situation. After this short conversation (inFrench), the pilot went quickly to the central control panel and put the engine-room telegraph to full astern. While the pilot was occupied with the VHF conversation and the engine-room telegraph, the OOW approached the steering console and twice toggled the steering mode selector (SMS) switch, leaving the switch at the HAND position. He then saw the rudder angle indicator begin to move to port. At some time during these events, the OOW put the wheel to midships. He then went to the ship's public address system and announced the steering problem. The first engineer, who was on the main deck close to the steering gear flat access door, ran immediately to this location upon hearing the announcement. As he entered the steering gear flat, he noticed the rudder was at the midship position. The time interval between the announcement by the OOW and the first engineer's arrival at the steering gear flat was about 15seconds. He quickly verified that both steering gear pumps were on. By this time, the ship's electrician had joined the first engineer in the steering gear flat and was assisting. In order to verify the steering gear operation locally, the ship's electrician activated the switch isolating the steering gear from the bridge. The first engineer then turned the rudder a few degrees to port and to starboard locally, using the trickwheel. The rudder responded to the trick wheel, so the bridge isolation switch was repositioned to give control back to the bridge. He then called the bridge on the sound-powered telephone and reported that the steering gear was functioning satisfactorily. The above sequence of events is summarized in Table1.5 Figure1. Approximate track of the Alcor from channel to grounding position By this time (1440), the ship was veering out of the channel on a heading of approximately 265T. The master had already re-entered the wheelhouse interior and was assisting the navigation team. The chief officer had come to the wheelhouse and quickly deployed to the forecastle to standby the anchor. Shortly afterward, the astern thrust became effective and the vessel slowed considerably. The pilot was informed at or about this time that the rudder was responding to helm action. By 1444, the vessel had effectively come to a stop. Engine movements ahead were attempted in the hope of reaching deeper water, barely one cable ahead, but to no avail. The starboard anchor was then let go. The vessel was now aground, at position latitude 4703'29.5N, longitude 07045'09.1W, on a heading of approximately 285T. (SeeFigure1 for the vessel track as recorded by the pilot's portable differential global positioning system [DGPS].) At 1444, the pilot reported to VTS that the vessel had left the channel but that they were not yet aground. At 1506, VTS asked the pilot if they needed assistance. The pilot responded that he thought the vessel could come free but would consult with the master. He also confirmed that the rudder was now working. At about this time, VTS inquired whether they required a speed reduction for vessels transiting the area. The pilot responded in the negative. At 1540, the pilot reported to VTS that they were still trying to extricate the vessel. Immediately after the grounding, the master contacted the owners to consult about tugs. About one and a half hours after the grounding, at 1615, one tug was ordered via VTS. The tug OceanCharlieleft Qubec at 1705 and arrived on scene at 1930, one and a quarter hours after high tide. Although on approximately the same heading of 285, the Alcor had moved 2cables to the southwest, pushed by the flood tide current, and had settled in water between four and six metres above chart datum, in position latitude 4703'18N, longitude 07045'33W. No attempt was made to refloat the vessel at this time, as the tide had already dropped by about one metre. After consulting with officials of the Corporation des Pilotes du Bas Saint-Laurent (Pilot Corporation)by cellular telephone, the pilot decided to remain on board and assist. By 1700, Transport Canada (TC) officials had arrived on board. Shortly after their arrival they conducted their initial steering-gear tests, which showed that the steering gear appeared to be functioning adequately and within the prescribed time limits.6 Soundings were taken throughout the ship. Hopper tanks 2and3 on the port side and hopper tank3 on the starboard side were determined to be taking on water. At 1752, the pilot of the Alcor asked VTS for a speed reduction for local traffic. A 1.3 Injuries to Persons No one was injured as a result of this occurrence. A 1.4 Initial Hull Damage Hopper tanks (water ballast) 2(S) and 3(P) were holed as a result of the initial grounding. The hull later sustained further extensive damage (seePartB1.3, HullFailure). A 1.5 Certification A 1.5.1 Vessel The vessel's Certificate of Maltese Registry, Certificate of Class, International Load Line Certificate, Construction Certificate, Safety Equipment Certificate, and Radio Station Certificate were valid and appropriate to the service in which she was engaged. Classification-related inspections and surveys of the Alcorwere carried out on behalf of the owner by the Russian Maritime Register of Shipping (RS). International regulatory and national registry certification-related inspections were carried out by RS on behalf of, and under the authority of, the Government of the Republic of Malta. The safety management systems of both vessel and management were audited by RS, and found to be in accordance with the requirements of the International Safety Management Code (ISM Code). The vessel's Safety Management Certificate was valid until April 2003; the managers' Document of Compliance was valid until March 2003. A 1.5.2 Personnel Certificates of competency for the master and officers were valid and complied with the provisions of the International Convention on Standards of Training, Certification, and Watchkeeping for Seafarers.The certificates were appropriate to the service in which the vessel was engaged. Qualifications of the crew were in accordance with regulatory requirements. The master possessed a Master Mariner certificate, issued on 15May1992 in Ukraine. The OOW possessed a certificate of an officer in charge of a navigational watch, issued on 14March1995 in Ukraine. The pilot possessed a Master Mariner certificate, issued on 05May1994. He had acquired his ClassC pilotage licence on 01April1998. Following a regulatory change that had come into effect 20days before the accident, he was entitled to a ClassC-1 licence (ships up to 30000deadweighttons(DWT)) with the coming into force of the amended regulation on 21October1999. His pilotage licence was thus amended to this higher tonnage on 02November1999. A 1.6 Personnel History The master had been at sea for 39years, the last 10years as master on various ships. He had been on the Alcorsince 09April1999. The OOW had been at sea on various ships as an OOW since May1995. He had joined the Alcor on 09April1999. The pilot had begun his sea experience as OOW in 1989. In 1992, he obtained his Master Home Trade certificate and his Master Mariner certificate in 1994. Prior to entering the Pilot Corporation, he served on ocean-going vessels as OOW for a total of 45months' sea-time. He began his pilotage apprenticeship on 01April1996 and was granted his pilotage licence on 01April1998 having completed 226trips on vessels of all sizes as an apprentice pilot. Since 01April1998 and until the Alcorassignment, he had performed approximately 175pilotage assignments. It was his first experience on this vessel. A 1.7 Weather, Current and Tide A 1.7.1 Wind, Seas and Visibility At the time of the initial grounding, on the afternoon of 09November1999, winds were 10to15knots, generally from the northeast. Visibility was good and seas were calm. A 1.7.2 Current at the Time of Grounding The most significant currents in the St. Lawrence estuary between Trois-Rivires and the Saguenay River are the product of tidal forces.7 In the area of the grounding, the current changes direction from a southwesterly (upstream) flow with the flood tide to a northeasterly (downstream) flow with the ebb tide. This change in direction comes about gradually, starting about 50minutes after high water is reached at this point. The maximum force of the tidal current varies with the tidal range, and can exceed 3knots. At the time of the grounding, the tidal current was setting approximately 215T at between 2.5and 3.5knots. A 1.7.3 Tide In the St. Lawrence River, tidal forces reach as far upstream as Trois-Rivires, diminishing to nearly zero on Lac Saint-Pierre.8 The tides are a mixture of diurnal and semi-diurnal, although the semi-diurnal oscillation dominates. The result is a semi-diurnal cycle that differs in height and duration from one cycle to the next. Although the maximum tidal range is experienced at le-aux-Coudres (7.1m), very large ranges are experienced as far upstream as Qubec (5.8m). The reference Port of Saint-Franois, led'Orlans, some 4nm upstream of the grounding site, has a maximum tidal range of 6.6m.9 The measured tidal values, in metres, for this reference port on November09, 10,and11 were as follows: The time of the course alteration off buoy K-108, 1437, was 1hour and 37minutes after the measured low tide of 0.55m. The measured tidal height at this time was 2.2m above chart datum. The static under-keel clearance of the Alcorin the dredged channel at this time and location was, therefore, approximately 6.7m. The measured rate of rise of the tide at this time was in the order of 1.3m/h./ A1.8 Bathymetry and the Navigable Channel - Traverse du Nord A 1.9 Navigation and Steering Equipment A 1.9.1 Navigation Equipment The Alcoris equipped with all electronic aids to navigation required by international conventions, including X- and S-band radars, supplemented by a global positioning system receiver. Additionally, the pilot had brought on board a portable navigation system that plotted the vessel's progress along the navigable channel. This system, a Starlink DGPS with a laptop display unit, does not show details such as coastline or soundings, but simply plots the vessel's position with respect to the buoyed, navigable channel (seeFigure2). Figure2. Two possible information displays from Starlink DGPS It also produces information such as course over-the-ground (COG), speed over-the-ground, and distance from the centre of the channel. Satellite information, DGPS correction, and other system health diagnostics can also be displayed. This system was designed specifically as a pilotage aid and was being used on a test basis during the voyage from Les Escoumins pilotage station. Although the pilot glanced from time to time at this equipment, he was using the ship's radars and visual leading lights, in the usual fashion, to conduct the Alcorup the river. A 1.9.2 Steering Gear The Alcorwas equipped with a four ram Mitsubishi electro-hydraulic steering gear (Rapson Slide type). There are two Mitsubishi Janney pump units to pump hydraulic oil, each driven by a 15kilowatt, 1800r/min electric motor. One or both pumps can be put on line when operating the steering gear. With one pump operating, 65 of rudder movement takes an average of 27seconds; with both pumps, it takes an average of 21seconds. The gear can be operated by a local trick wheel in the steering gear flat or from the helm unit on the bridge. The steering gear was designed to produce a torque of 50tonne/metres (t/m) at the maximum working pressure (170kg/cm2). A 1.9.3 Bridge Helm Unit Figure3. Helm unit of Hokushin steerin control system The Alcorwas fitted with a Hokushin steering control system, model PT-7J2. This system, fitted during construction of the vessel and approved by the classification society, includes an all-electric helm unit, incorporating hand-steering capabilities and autopilot functions. One switch selects either autopilot or hand steering and is located on the lower left-hand side of the steering unit front piece (seeFigure3,frontview, switchNo1).10 If hand steering is selected on switch No1, a second, three-way SMS switch (No2), labelled Pilot Main on this unit, permits the choice of OFF, HAND, or NFU. Switch No2 is located on the lower right-hand side of the steering unit front piece, 55 mm behind the spokes of the steering wheel. The detent torque required to turn the SMS switch was found to be 0.6 newton/metre (Nm).11 No quantitative norm or regulation governing detent torque is known to exist for this type of switch in this type of application. When the SMS switch is in the OFF position, turning the wheel or moving the NFU lever one way or the other does not impart a signal to the steering gear. With this switch in the HAND position, a signal is sent when the wheel is turned to the left or to the right - turning the wheel two full rotations is equivalent to 35 rudder angle. When the switch is in the NFU position, the spring-loaded NFU lever must be held to one side or the other for a signal to be sent to the steering gear. Turning the wheel when the switch is in the NFU position - an independent mode - does not impart a signal to the steering gear. Steering commands from the bridge helm unit are transmitted through one of two independent wiring paths, each connected to its own solenoid valve (autopilot directional valve) at the steering gear. Signals are also amplified by one of two independent amplification arrangements. Selections are made at the bridge helm unit as to power unit 1 or power unit 2 (wiring path), and/or amp 1 or amp 2 (amplification unit). This arrangement can be interswitched in any combination and serves to provide backup in the event of any one of the systems failing. Both these switches were reported to have been in the No. 1 position at the time of the grounding. A1.10Tests and Surveys of Steering Gear and Steering Control System At approximately 1700 on 09 November 1999, TC inspectors boarded the Alcor to assess the situation and verify the seaworthiness of the vessel. Steering gear failure had been initially reported, so the inspectors surveyed the steering gear upon their arrival. Since visual inspection revealed no anomalies, inspectors carried out operational tests from the wheelhouse and the steering gear flat. The steering control system and steering gear performed without any apparent fault, although port-to-starboard operation was marginally longer than starboard-to-port operation. This may have been attributable to obstructions on the river bottom at this time, as the vessel was aground. Similar surveys and tests were carried out by TSB on 20November1999. No anomalies were found, and times were similar for all directions of operation. On 15 December 1999, a complete verification of the electric components of the steering control system, including the steering station, wiring connections to the steering gear, and electrical components at the steering gear were carried out. No anomalies were found with the electrical system. On 16 December 1999, certain hydraulic components of the steering gear (the steering pumps, relief valves and directional valves) were bench tested at a shore facility. Results showed that the steering pumps, fitted at the time of construction, had a flow capacity approximately 17%below original specifications. Solenoid valves were found to be operating normally, although one of these valves, powerunit2, had an infiltration of hydraulic oil within its casing. The starboard relief valve was found to be leaking at low pressure and approximate crack pressure was found to be 124.14bars (1800pounds per square inch [psi]). The port relief valve was operating at 158.6bars (2300 psi). The design pressure for these valves is 168.9bars (2450 psi). The hydraulic components that were left in place on board the Alcor, such as the rams, connecting piping, hydraulic oil reservoir and various valves, were surveyed. The main rams showed normal wear and scoring and the packing glands were seen to be leaking. The hydraulic oil in use at the time was tested and found to be more viscous than the manufacturer's recommendation.12 The oil contained trace elements of copper, iron and lead, indicating wear in the system components. All other components were found to be normal. Close examination of the steering gear arrangement and rudder angle indicator mechanism revealed no anomalies. When power system 1 was selected, there was no difference between actual rudder angle and rudder command. When powersystem2 was selected, there was an approximate two to three degree difference between actual rudder angle and rudder angle command in the wheelhouse, such that when the wheel was in the midship position, the rudder was approximately two degrees to starboard. A1.11 Pilotage Licences - Laurentian Pilotage Authority A 1.11.1 Current Structure Within each district, there are different levels of pilotage licences (ClassA through ClassD) based on criteria prescribed in LPA regulations. Limiting factors that define the different classes are the size of the vessel piloted and time worked at a particular level (with a minimum number of pilotage assignments). For districts1-1 and1, size limits of the vessels for various classes are determined by length of the vessel piloted. In District2, it is deadweight tonnage of the vessel that defines maximum size of the vessel for a given class. Table3 outlines these factors. ClassD licences are for apprentice pilots in all districts and for any size ship in the presence of a licensed pilot. A1.11.2 Evolution of LPA Regulations Regarding Pilotage Licences In District 2, the LPA regulations with respect to the vessel size limits has often been the result of proposals made by the Pilot Corporation.13 These proposals are often based on a popular vote within the Corporation membership. The LPA - as the legal authority entrusted with the mandate of administering, in the interest of safety, an efficient pilotage service in these areas - can accept or reject these proposals. Evolution of the criteria defining classes of pilotage licence in District2, both DWT and duration worked, began in 1983. In that year, the limiting factor criteria for size of vessel was changed, by regulatory amendment, in two ways.14 First, the net registered ton (NRT) unit of measurement was changed to DWT. Although the ratio of DWT to NRT can vary widely among vessel types, a sampling of river traffic in the St. Lawrence has shown the ratio to be somewhere between 1and3.5, with a typical average somewhere near2.7. Given this, the original limits as expressed in NRT for ClassB andC (10000and5000,respectively), would have been loosely equivalent to 25000DWT and 12500DWT, respectively. The second change brought about by the 1983 amendments was the increase of ship sizes to 50000DWT for ClassB and to 15000DWT for ClassC. In 1992, time spent as an apprentice pilot in District2 was reduced by regulatory amendment from three years to two years, although the minimum number of pilotage assignments remained approximately the same, such that the apprentices performed as many trips in two years as previously in three.15 In 1994, the DWT limit for ClassCpilots was increased from 15000DWT to 20000DWT. The accompanying Regulatory Impact Analysis Statement for the 1994 regulatory change, as published in the Canada Gazette, PartII, states, in part: as the number of mid-sized type vessels has decreased in recent years in that district. Therefore the amendment will permit holders of licences and Class C pilotage certificates to complete their training more effectively while under this Class.16 In 1999, the DWT limit for Class C pilots was increased once again (from20000 to 30000DWT). The following Regulatory Impact Analysis Statement for the 1999 regulatory change, as published in the Canada Gazette, PartII,states: The use of larger ships in these waters prompted changes to subsections 15(4) and(5) of these regulations. These provisions amend the current ship limitations for holders of ClassB and Clicences and pilotage certificates, based on a ship's length and deadweight tonnage. This initiative provides ClassB and ClassC holders with the experience of piloting more ships and larger ships which were previously piloted by more senior Class holders. In addition to enhancing the knowledge and training base for these Class B and C holders, this provision provides the Authority with greater flexibility in the dispatching of its pilots, thereby improving service efficiency, particularly in peak traffic periods.17 All of the above-mentioned regulatory changes are summarized in Table4. Minimum entry-level competency for pilots in District 2 was also increased in this period. Prior to 1980, a First Mate Home Trade, or Second Mate Foreign Going certificate was required as a minimum for District 2 pilots. A 1980 regulatory amendment made Master Home Trade or First-Mate Foreign Going certificates the minimum requirement.18 A1.11.3 Evolution of Service Agreements Regarding Pilotage Licences Within District 2, where the Alcor grounded, there are subclasses for Class B and Class C licences that are defined within the service agreement between the LPA and the Pilot Corporation. The service agreement respects LPA regulations for maximum DWT and duration worked and imposes lower DWT limits for the first years worked in a given class. The evolution of service agreements and the impact these agreements have had on the application of the regulation are summarized in Table5. The 2000-2003 Service Agreement reflects changes to LPA regulations that came into force on 21October1999. Changes include 30000 DWT as a limit for ClassC licensed pilots, up from the previous 20000DWT. The service agreement limits the first year worked as Class C to 20000DWT and then allows for the increase to 30000DWT in the second year worked. Also, by virtue of the service agreement, ClassB-3 has been eliminated and ClassesB-2andB-1 are for three years each as opposed to the previous two years. A 1.12 Pilotage Training A 1.12.1 Apprenticeship Program Pilots accepted into the program in the Laurentian Districts must undergo a two-year apprenticeship, during which they participate in pilotage of vessels in the presence of a licensed pilot. During these training trips, pilots can make written comments as to the adeptness or deficiencies observed in the apprentice's ability, but no formal guidelines exist for this process. Apprentice pilots must also undertake formal classroom instruction, in their first and second years, concerning knowledge of the district and general aspects of the profession, including low under-keel performance of vessels. At the end of the apprenticeship program, extensive oral and written exams are given. Vessel performance with low under-keel clearance may, or may not be, part of the series of theoretical questions asked during the exams. LPA regulations stipulate a minimum number of pilotage assignments (113) to be made each year as an apprentice pilot, although the Pilot Corporation's training plan is more rigorous, specifying 120trips. Harbour movages are also quantified for the major harbours of District2 (forDistrict2licences). LPA regulations do not qualify the type or size of vessel to be piloted, nor do they specify a minimum number of assignments to be made on ships with low under-keel clearance. On the other hand, the Pilot Corporation's training plan stipulates a minimum of three trips per year on vessels with draughts of 12.8m or more. According to the Pilot Corporation, at least 25% of all pilotage assignments by apprentices are on vessels having draughts of 10m or more. This is not a requirement, however, but a reflection of the size of vessels trading in the area. In order to evaluate the apprentice's actual performance, the Pilot Corporation's training plan stipulates that an apprentice must make at least two trips, during the second year of training, in the company of a Pilot Corporation board member.19 Evaluations are performed informally by observing the apprentice's pilotage skills during the trip. No formal evaluation process or tool is used during the accompanied trips and the apprentice's skill and knowledge of low under-keel clearance is not necessarily evaluated at this juncture. A 1.12.2 Post-Apprenticeship Training During their careers, pilots are offered additional training, including courses in bridge resource management (BRM) and ship-handling. All training deemed necessary by the Pilot Corporation is submitted to LPA for approval. Costs for training are covered off by a fixed amount allocated by LPA while the balance, if any, is assumed by the membership of the Pilot Corporation. The amount allocated for training is negotiated by the two parties and stipulated in the service contract. This amount has been unchanged since 1990. For the years 2001, 2002 and2003, it was agreed to increase this amount to cover the costs of BRM training. BRM training has been identified as a priority and will be mandatory for all pilots by 01January2005. Starting in 1972, the Pilot Corporation has been sending its pilots on a ship-handling course. Since the 1980s, this training has been a one-week course that employs working, scale model vessels that trainees manoeuvre through waterways that recreate currents, bank effect and similar phenomena. The models are used to teach practical aspects of ship manoeuvring. Classroom time is devoted to ship-handling theory, including subjects such as the dynamics of a vessel's pivot point. Initially, this course was offered only to ClassA pilots. Beginning in 1990, it was extended to ClassB pilots. In 1993, it was enshrined in the service agreement with LPA, such that all ClassB pilots would be sent on this course in the first year of their service as a ClassB. As of the end of 1999, all ClassA pilots in District2 had taken the course (95% have taken it twice), as had all ClassB pilots, save one. No ClassC pilots had yet received this training. Because of the high cost of such training and the allocation of funds by the LPA, the Pilot Corporation could only assign a maximum of 12participants per year to the course. If more than 12new ClassB pilots are entitled the training, participants are selected by seniority. In other pilotage areas of Canada, candidate selection for this training is different. The Central St.Lawrence Pilot Corporation (LPA District 1) selects candidates for this training at random amongst all classes of pilots. In British Columbia, the Pacific Pilotage Authority sends all pilots on the ship-handling course during the fourth month of a six-month apprenticeship program. The Great Lakes Pilotage Authority does not, as yet, send its pilots on such training, but a recent review of training requirements has identified it as a need. Figure4.Area chart with grounding positions Part B Salvage B 1.0 Factual Information B1.1 Preparations for the First Refloating Effort B 1.1.1 Loading and Trim The Alcorloaded pelletized foundry clinker at Le Palito, Venezuela, on 25October1999. The vessel was repositioned several times during the loading operation to provide for the clear flow of cargo from the loading facility into each cargo hold. This ensured satisfactory distribution, loading rates and hull stresses. The final distribution of cargo was such that holds1, 2, 4, and5 were partially filled, while hold3 remained empty. The distribution of cargo was similar to several of the typical loading conditions included in the vessel's approved stability and strength booklet (LoadingManual). Total cargo deadweight was slightly less than that of the nearest comparable loading condition (in which hold3 also remained empty). The lighter cargo deadweight resulted in lower sheer forces and still water bending moments (SWBM) imposed on the hull; on departure, the SWBM was approximately 40% of the approved maximum. Once loaded, the recorded draughts were 9.77m forward and 9.86maft. B 1.2 First Refloating Effort As previously mentioned (SectionA1.2.2), one tug was ordered some 1.5hours after the initial grounding, and it arrived on scene at 1930, 09November1999. After high tide, refloating the vessel was not possible, and one tug was found to be insufficient to the task. Due to the short period before the next high tide, the unavailability of transhipment vessels, and water depth restrictions around the Alcor, no lightering operations were planned. The Alcor did not have self-unloading capabilities, so sacrificing cargo to reduce draught was not an option. Early on the morning of 10 November 1999, loud reverberations were heard throughout the ship. Small cracks were discovered on the main deck, on the starboard side at frame120, and on the port side between frames95 and100. The master, TC officials, and the pilot agreed that the Alcorshould be refloated as quickly as possible, as huge strains were being imposed on the structure with each low tide. The river bottom forward was approximately two metres lower than that aft of amidships. By the evening of 10November1999, some 28hours after the grounding, a salvage effort was made with four tugs. A Lloyd's Open Form was agreed to only just before the refloating manoeuvre. By 1745, hopper tanks2 and3 on the port side and hopper tank3 on the starboard side had been pressed with compressed air. Although the pilot had suggested the manoeuvre proceed to the north (moving ahead), the plan the salvors used was to pull the Alcor astern and into deeper water in a southeasterly direction. By 1755, the four tugs were in position. Three unsecured tugs on the port side pushed to keep the vessel from riding up higher onto the bank with the flood tide, while one, secured astern, pulled in a southeasterly direction. By 1815, the Alcorwas pivoting about her centre, approximately between the headings of 285T and 055T. At 1835, one of the three pushing tugs was sent forward, secured, and pulled in concert with the stern tug. At approximately 1900, nine minutes after the measured high tide of 5.96m and with the flood current decreasing in strength but still in a southwesterly direction (220T), the Alcorbegan to move astern under tow and her own power. Figure5 Initial (1) and second (2) grounding positions (all positions approximate) Soon after the Alcorbegan to move astern, the tugs stopped assisting under the salvage master's instructions, while the engine of the Alcorwas kept moving astern for approximately two minutes. During this time, the vessel's heading was fairly constant at about 285 T. Soon after the vessel started astern, the salvage master inquired of the pilot, who was at one of the radars, if the vessel was in safe water. When the pilot responded in the affirmative, the salvage master put the engine of the Alcorto stop; the salvage master had not yet handed the con over to the pilot. Shortly after this, the ship stopped moving astern. The engine was again put to full astern by the salvage master and the tugs resumed their assistance, but to no avail. The Alcor had moved some 2.8cables to the southeast and had grounded for a second time at position 4703'08N, 07045'12W (seeFigure5). By 1945, the falling tide had dropped by about 0.5m. Despite the continuous effort of the tugs, the vessel remained immobile. A decision was taken to suspend the salvage operation until the following high tide. At about 2200 that evening, the pilot requested a relief break from the pilotage dispatch centre in Qubec, some 31hours after the initial grounding. Photo2. Starboard side fracture B 1.3 Hull Failure At approximately 0015 on 11November1999, a loud reverberation was heard throughout the ship. A large fracture had developed transversely across the main deck, near frame110 on the starboard side, through the No4 hatch coaming, and across to frame87 on the port side. The fracture extended down both sides and stopped just short of the hopper tanks. Holds3 and 4were opened to the sea and the fracture on the main deck was as wide as 0.52m in some places (seePhotographs2,3,and4). For safety reasons, all but a skeleton crew were evacuated from the vessel. The salvage company abandoned the Lloyd's Open Form salvage agreement and relinquished control of the vessel. Later that day, the owners of the Alcor were served a request of intention on behalf of both the Canadian Coast Guard (CCG) and TC under the auspices of the Navigable Waters Protection Act and the Canada Shipping Act, respectively. The owners were instructed to present a plan for removal of the vessel in short order or risk loss of control of the process. Photo4 Main deck fracture (at arrow) Time was of the essence, as winter ice conditions could aggravate the salvage operation and further compromise the vessel's structural integrity. First ice formation was predicted for as early as 13 December 1999. B 1.4 Second Refloating Effort B 1.4.1 Preparations On 19 November 1999, a second salvage company was selected. The refloating effort was scheduled for the evening high tide of 07 December 1999. As this date approached, unfavourable winds were predicted for both December06 and07. The refloating was advanced to the afternoon high tide of 05December1999. By then, preparatory work for refloating had been completed, which included an underwater survey of the hull, ballasting down, strengthening in way of the fracture, removal of unnecessary fuels, lightering of cargo (approximately 11200t had been taken off), and a detailed hydrographic survey of the grounding area. Three special buoys had been placed nearby to indicate the deep water limits in the vicinity of the vessel. Closing the Traverse du Nord during the refloating and transit of the Alcorthrough this section of the river had been discussed informally by TC, CCG, and salvors. However, no explicit plan or directive was in place to execute this action. On the morning of 05 December 1999, the salvage master briefed officials from CCG and TC, as well as the refloating team, including the tug masters, and the owners' representatives, on the planned manoeuvre. The plan was to move the vessel forward once afloat, and then turn her stern to starboard in order to back her out of the confined area bordered by shoals on either side. Once in safe water, the con was to be passed to a river pilot for passage to Qubec. The four river pilots scheduled for the operation, two for the Alcor and two for the lead tug, were not present at this time, but were briefed upon their arrival on board a few hours later. Although no specific location was indicated at which the hand-over from salvage master to pilot was to take place, the pilots were informed that they would be handed the con once the vessel was safely in the channel. Figure6. Refloating manoeuvre (SB arespecial buoys used for the refloating; positions approximate) B 1.4.2 Refloating Manoeuvre The Alcorhad come afloat by 1515 and two unsecured tugs were used on the port side to keep the vessel on station. One tug had been secured forward and another secured aft. At this time, the refloating team, TC, and CCG conducted a survey to assess the structural integrity of the vessel before moving it into deeper water. The gyro compass was now unreliable and could not be used. The magnetic compass was also unreliable, due to the considerable amount of reinforcing steel brought on board to strengthen the broken vessel. Although the vessel's radars were operational, they had to be used in the ship's head-up configuration due to the unreliable gyro input. In the wheelhouse at this time, apart from TC and CCG officials who were observing the operation, the bridge team consisted of the salvage master, a salvage captain (acting as master), and the two river pilots. There was no dedicated helmsman or OOW. Figure7. Refloating into channel (positions are approximate) At approximately 1540, under her own power and with the help of the tugs fore and aft, the Alcorwas first moved forward (seeFigure6, position1) and then pivoted, the stern moving to starboard, until the vessel was on an approximate heading of 215T, parallel with the shoals on either side (seeFigure6,position2). Once in this orientation, sternway was put on using the stern tug and the engine of the Alcor. The operation was executed as initially planned. At approximately 1550, with the bow of the Alcorjust past the special buoy to port (seeFigure7, position3), the salvage master asked if one of the pilots was ready to assume conduct of the vessel. The first pilot replied in the affirmative and took the con, asking the stern tug to pull in the direction of the K-108. The second pilot was using one of the radars and the salvage captain the other. Although visibility had been good before the refloating, it had now diminished to approximately 0.5nm, and even less at times. The Alcorwas still going astern when the salvage captain, who was also plotting positions on the chart, declared that the vessel was coming dangerously close to the shoal named Le Banc de Sable, 1.2nm south-southwest of Cap Tourmente. Although the second pilot, who was verifying the vessel's position on the radar, was sure the vessel still had plenty of sea room, the first pilot, who had the con, ordered the stern tug to stop pulling and ordered ahead engine on the Alcor. The salvage master put the engine-room telegraph to half ahead (seeFigure7, position4). B 1.4.3 Change in Plans During the astern manoeuvre, the Alcorhad inadvertently been allowed to turn to port, from an initial heading of 215 T to somewhere between 140T and 150T. There are differing views as to why this happened, but it is generally accepted that there was an unwanted pulling in the forward tug's towline as the astern manoeuvre was carried out. Ideally, this line would have remained slack throughout the astern manoeuvre. Later, with ahead thrust from the Alcorand the forward tug now pulling, the Alcor stopped moving astern and began to move forward. Shortly thereafter, at approximately 1615, out of the mist ahead and to starboard, the buoy SB3 became visible from the wheelhouse of the Alcor(seeFigure7, position5). The second pilot, who had been assisting, became more assertive in the absence of any orders from the first pilot. He ordered the lead tug to pull the Alcorin the direction of K-108, in order to distance the vessel from the shoal water to starboard. This action further moved the bow of the Alcor to port, to a heading of between 110 T and 100 T as the vessel entered the channel (seeFigure7, position6). As she came into the channel, the Alcorwas now headed more downriver than upriver. The second pilot, who had by now assumed de facto conof the vessel, decided to continue downriver and turn at Sault-au-Cochon, where more sea room was available to manoeuvre. By 1617, the vessel was straightened out in the channel and headed downriver. B1.5 Damage Subsequent to the First Refloating Effort During the early morning hours of 10November1999, near low tide, small cracks on the main deck were observed and recorded by the crew: near frame120 on the starboard side and between frames95 and100 on the port side. The river bottom at this initial grounding position was such that the aft 40% of the vessel was in approximately four metres of water while the fore part of the vessel was on a gradual slope to deeper water with the bow in about six metres of water.31 Orientation of the vessel in the second and final grounding position, shortly after 1900 on 10November1999, and its footprint on the riverbed were such that the bottom shell plating aft of midships was in approximately 2.5m of water. This part of the vessel thus maintained effective bearing contact throughout the tidal cycles. The forward half of the vessel was on a gradual slope, with the bow in about five metres of water. The buoyant support of the forward half of the grounded vessel fluctuated as the water level rose and fell with the tides. The maximum tidal range prior to the main structural failure of the Alcorwas 5.31m, which was slightly more than half of her original forward draught when afloat. Once settled in the final grounding position on 10November1999, conditions that created very high bending moments and tensile stresses in the upper members of the hull girder included: a large reduction of buoyant support during the low tides, a loss of intact hull buoyancy due to flooding of breached hopper ballast tanks 2 and 3 (P), and 3 (S), and a deadweight of cargo in holds 1 and 2 in the unsupported forward end of the vessel. The SWBM near the mid-length of the hull was much greater than the approved maximum SWBM related to the vessel when free floating, and eventually exceeded that which the main deck structure of the grounded vessel could withstand. Tensile stress concentrations at minor discontinuities in the upper members of the hull girder initiated brittle fractures in the main deck plating that subsequently propagated across the deck, into the deck longitudinals, then through the sheer strake and gunwale, finally propagating down the side shell plating. Principal fractures in the port and starboard side shell plating breached the watertight integrity of cargo holds3 and4 and upper wing water ballast tanks2(S) and3(P). The width of the fractures across the main deck plating near midships widened to approximately 0.52m. The longitudinal integrity of the hull was only maintained by the bottom shell plating, the inner bottom tank top plating, and the internal double bottom structure. These lower members of the hull girder were subjected to compressive stress loading and remained intact, collectively acting in the manner of a large hinge. B 1.6 Damage to the Environment The grounding was within 0.5nm of a bird sanctuary on the banks of the St.Lawrence River. Also, the area is the natural habitat for several species of duck. Approximately 25t of clinker spilled out of holds3 and4 into the St.Lawrence River. Clinker is not considered a marine pollutant32 and the spill was found to pose no risk to the bird or fish habitat. No heavy fuel, diesel or other marine pollutant was released into the environment subsequent to either the grounding or the rupture of the ship's structure. B 1.7 Weather, Current and Tide On the evening of 10November1999, at the time of the first refloating effort, winds were 15to 20knots from the northeast. Visibility was at times reduced by snow, but on the whole remained good. The tidal current at 1900 was in a southwesterly direction, setting approximately 220T at between 0.5and 1knot. On 05 December 1999, during the successful refloating, visibility was reduced at times to less than a one nautical mile due to mist, particularly during the actual refloating manoeuvres between 1530 and 1630, where at times it was reduced to less than 0.2nm. Once the Alcor was in the channel and underway, visibility was reported as good. At 1600, as the vessel was moved into the channel, the tidal current was setting approximately 220T at between 0.5and 1knot. Winds were calm at this time. B 1.8 Governmental Infrastructure The waters in question are within VTS jurisdiction. The VTS mandate, other than communications, is generally limited to traffic advisories and information. Under special circumstances, VTS can direct traffic. (SeesectionC1.5 of this report for details on traffic direction.) Another division of CCG was quickly involved in the Alcor incident: the Environmental Response Division of the regional Coast Guard / Marine Programs Directorate. As per their mandate, they monitored the situation for environmental considerations. Transport Canada Marine Safety (TCMS) was active during the period that the Alcorwas aground. TCMS surveyors were aboard the vessel within hours of the grounding. After the hull fracture, they remained on board on a continuous basis to monitor the vessel's condition. Once the second refloating effort was underway, they re-evaluated the structural integrity of the vessel before allowing the transit to Qubec. The Navigable Waters Protection Act provides the Minister with the necessary powers to remove a stranded vessel if the difficulty or danger continues for more than 24 hours.33 B 1.9 Pilot Relief (District 2) For District 2, paragraph 35 of the LPA regulations stipulates that two pilots are assigned to a vessel when any one of the following conditions are met: the ship is likely to be under way for more than 11 consecutive hours in that district; on a ship in excess of 74 999 DWT; on a tanker of 40 000 DWT or more; on a passenger ship of more than 100 m in length, or on any ship during winter navigation. The requirement helps reduce risk in two ways. First, for long assignments, the two-pilot requirement acts as a fatigue countermeasure. Each pilot will work a mutually-agreed duration, and is then relieved by the second pilot, and vice versa. Second, for particular vessels where the potential severity of consequences has been seen to justify using a team approach, the requirement serves to eliminate single-point failure. In these instances, although each pilot works in turn for the majority of the voyage, a team approach is adopted for passage of strategic areas such as the Traverse du Nord. There are no written guidelines or work procedures, nor is it specified in the regulations when or if the two pilots must work as a team; this is left up to their discretion. Additionally, there are no set criteria established for pilot assistance or relief in emergency situations when one pilot is on board. Under these circumstances, the current LPA and Pilot Corporation procedures place the onus on the pilot involved in the occurrence to make a decision as to his/her relief. B1.10 International Safety Management Code Procedures Section 8 (Emergency Preparedness) of the ISM Code specifies the following: The company (manager) should establish procedures to identify, describe and respond to potential emergency actions. The company should establish programmes for drills and exercises to prepare for emergency actions. The safety management system should provide for measures ensuring that the company's organization can respond at any time to hazards, accidents and emergency situations involving its ships. B 1.10.1 Vessel Procedures (Owners/Operators) Emergency preparedness procedures were contained in the vessel's shipboard operations manual (section6of volumeII). Chapter2 of this section contained advice on casualties including, among others, stranding, steering gear failure, flooding, salvage, and hull failure. AppendixQ (Salvage) of the manual also states: In most cases, where time and circumstances permit, the owners together with the master will agree terms with the salvors on which salvage services will be rendered with the authority of other parties who have interest in the vessel and may benefit from salvage services. Therefore in the event salvage services are required it is important that the master informs the company as soon as [the] casualty occurs to prevent salvage services becoming more urgent and consequently more expensive. However, in cases of absolute urgency, the master himself may negotiate the terms of the salvage agreement with the salvors, normally Lloyds Open form. It must be stressed that the master only has authority to reach an agreement in cases where the vessel and the cargo onboard are in imminent danger and there is no reasonable opportunity to contact owners and cargo owners and any other party with an interest in the vessel who will benefit from the salvage services in order to obtain their authority. Part C Risk of Collision C 1.0 Factual Information C 1.1 Particulars of the Vessels Figure 10. Vessel positions as Eternityand CanmarPride pass 1.2.1 Vessels at Anchor No vessel was allowed to pass the Alcorwhile she was in the Traverse du Nord. Several downbound ships were held above buoy K-136, the upstream limit of the Traverse du Nord. Five vessels were at anchor in a 3.5nm stretch of the river between Pointe Saint-Jean and Rivire Maheu. The CanmarPridewas anchored closest to the Traverse du Nord, at the anchorage area known as Pointe Saint-Jean. This anchorage, although not formally indicated on marine charts, is locally accepted as an area south of Pointe Saint-Jean on le d'Orlans, far enough from traffic lanes so as not to impede or pose a risk to transiting vessels. When the CanmarPride came to anchor, only the name of the anchorage area was given, and VTS did not request a more accurate position. Information at hand shows that the CanmarPridehad anchored on the extended line with the leading lights of 053T, and at a distance of three cables from the intersection of the recommended routes 053/033. At anchor, the tanker Eternitywas the vessel farthest from the Traverse du Nord, having anchored approximately 3.5nm southwest of the CanmarPride. C 1.2.2 Other Vessels The CCG vessel George R. Pearkes had been tasked to escort the Alcor to Qubec; no other specific duties had been assigned this vessel. The upbound Algosar was permitted to follow the Alcorthrough the Traverse du Nord, without passing. C 1.3 Personnel History The pilots on board both vessels were ClassA pilots for District2; on the Eternity, the pilot had 21years' experience as a pilot in this district; on the CanmarPride, the pilot had 34years' experience in this same district. C 1.4 Weather, Current and Tide At 2300 on 05 December 1999, the time of the near collision, visibility was reported as good, with winds out of the west at 10knots. The tidal current was on the ebb, setting approximately 060T at between two and three knots. C 1.5 Occurrence Reporting Requirements The Transportation Safety Board Regulations, made pursuant to the Canadian Transportation Accident Investigation and Safety Board Act, the Shipping Casualties Reporting Regulations, made pursuant to the Canada Shipping Act (CSA), and the Laurentian Pilotage Regulations,made pursuant to the Pilotage Act, all require a near-collision to be reported immediately by the fastest means available, one of the methods being a report to the nearest shore station. This is to be followed by a written report to appropriate authorities. After the incident, the pilots and masters on both the CanmarPride and the Eternityall concurred that there had been a risk of collision, but neither vessel reported this incident to VTS by radio. There are indications, however, that officers within LPA were summarily apprised of the incident through unofficial channels. After some preliminary inquiries, the decision was taken within LPA to not investigate further. TSB was informed of the near collision on 21December1999. C 1.6 VTS in Canadian Waters VTS operates under the auspices of the Vessel Traffic Services Zones Regulations,pursuant to the CSA, and administered by the Marine Programs Directorate of CCG, Department of Fisheries and Oceans. The service is carried out by the Marine Communications and Traffic Services (MCTS) program of CCG. The objective of VTS is to protect the marine environment and improve the safety and efficiency of traffic movement, by providing the following services:48 a VHF traffic information and advisory service; a traffic clearance and screening service; a radar navigational assistance service; and a space management service, organizing ship movements in order to facilitate efficient traffic flow. Under the regulatory regime, no ship shall enter, leave, or proceed within a VTS zone without having previously obtained a traffic clearance49 and a report shall be made to a marine traffic regulator immediately before beginning and after completing a departure manoeuvre in a VTS zone.50 Part D Vessel Survey D 1.0 Factual Information D 1.1 Vessel Survey Requirements International Maritime Organization ResolutionA.744(18), Guidelines on the Enhanced Programme of Inspections During Surveys of Bulk Carriers and Oil Tankers, adopted on 04November1993 and subsequently incorporated as Chapter XI in the International Convention for the Safety of Life at Sea, formally came into force 01January1996. Under the Enhanced Survey Program (ESP), special surveys are conducted at five-year intervals by the vessel's classification society. In the case of bulk carriers, surveys become more rigorous as a vessel ages. Additionally, annual surveys are required, the second or third of which must be a more detailed intermediate survey. These survey reports may also incorporate memoranda that address and monitor specific ongoing items of concern arising from any preceding survey. For operational convenience, the scheduling of surveys is subject to some flexibility, (such that the intervals between routine annual and intermediate surveys may vary by plus or minus three months) while ensuring that actual elapsed time between special surveys is maintained substantially at five years. D1.2 Hull Survey and Inspection History - Alcor D1.2.1 Special Hull Survey In accordance with ESP requirements, and while named Mekhanik Dren, the (then) 20-year-old vessel was the subject of a special survey in 1997, while at Shanghai, in the People's Republic of China. Principal structural areas subjected to close-up visual inspection included: all cargo holds, side framing, inner bottom and transverse watertight bulkheads; internal structure of all upper wing ballast tanks; internal structure of all double bottom water ballast tanks; and main deck, side and bottom shell plating. Ultrasonic thickness gauging of the principal hull girder structural members was carried out in accordance with class requirements and recommended International Association of Classification Societies procedures. The inspections and thickness gauging resulted in extensive structural repairs and replacements, totalling 260tonnes of steel, mainly while the vessel was afloat. Principal structural repairs were in way of upper wing water ballast tanks1,2,3and4, port and starboard. Repairs in these tanks included renewal of several sections of main deck longitudinals, and repair or partial replacement of transverse webs, end bulkheads, main deck and wing tank sloping bottom plating. Internal inspections showed that cathodic protection systems were not fitted in any of the vessel's water ballast tanks. Double bottom water ballast tanks (1,2and3,portandstarboard) were fitted with a protective coating, which was recorded as being in fair condition. All upper wing water ballast tanks were recorded as being uncoated. Concluding remarks of the hull survey report included a notation: due to no protection or coating in topside water ballast tanks, it will be necessary to carry out common survey for it during next Annual Surveys. Hull, machinery, and electrical enhanced surveys, together with related repairs were completed to the satisfaction of the attending class surveyors, and ESPwas added to the vessel's Russian Maritime Register of Shipping (RS) class notation symbol. The special hull survey was completed and approval assigned in accordance with RS class requirements on 20June1997. D 1.2.2 Routine Annual Hull Survey (1998) The first routine annual hull survey was carried out in 1998, while the Alcorwas afloat in the Port of London, United Kingdom. The survey included general examination of the hull, forepeak and afterpeak tanks and cargo holds, as well as close-up inspection of two forward holds and attention to any newly incurred structural damage or ongoing, structure-related notations of previous surveys. No newly-incurred structural damage or substantially corroded areas were noted at this time. The absence of cathodic protection systems in any of the vessel's ballast tanks was again recorded, and the condition of the internal coatings in double bottom water ballast tanks was reported to be fair. The routine annual hull survey was completed to the satisfaction of RS class surveyors, and approval assigned on 19March1998. D 1.2.3 Port State Control Survey In accordance with the Canadian Bulk Carrier Inspection Regime and Port State Control (PSC) requirements, the Alcor, as a dry bulk carrier more than 15years old, was inspected while at Vancouver, British Columbia, on 25September1998. The inspection showed that an oily water separator bypass was installed in the vessel, which contravened regulatory requirements of the International Convention for the Prevention of Pollution from Ships. The master was informed and RS was advised of this contravention. The report called for the earliest implementation of remedial action to ensure both regulatory compliance and RS approval. No internal or close-up structural inspections were carried out at this time, and the vessel was not detained. The Alcorwas scheduled to undergo a PSC inspection upon arrival at Trois-Rivires, Quebec, with the inspection to include visits to the forepeak, afterpeak, upper wing tanks and cargo holds. D 1.2.4 Routine Annual Hull Survey (1999) A second routine annual hull survey was carried out while the vessel was afloat in the port of Bombay, India, in 1999. This survey included general examination of the hull, hatch covers, coamings and fittings, cargo holds, and forepeak and afterpeak water ballast tanks. The lower 25% of the side framing and adjacent shell plating in the cargo holds, and all of the internal structure of each of the upper wing water ballast tanks were subjected to a close-up survey. On general examination, the hull, hatch covers, coamings and fittings, all cargo holds, forepeak and afterpeak water ballast tanks were found to be satisfactory. A close-up survey showed the lower side framing in all cargo holds and the internal structure of all upper wing water ballast tanks to be in satisfactory condition. It was also recorded that there was not found substantially corroded areas. The protective coating in the forward and afterpeak water ballast tanks was reported to be in poor condition, while that in all the cargo holds was reported as good. The absence of cathodic protection systems or protective coatings in any of the upper wing water ballast tanks was also recorded. Concluding remarks of the report included the notation: Due to no protection/coating in topside ballast tanks Nos1,2,34, (PS) it will be necessary to carry out measurements of thickness of sloping plating and bulkheads in tanks, as well as close-up survey of framing in topside tanks, fore peak and after peak for it during Intermediate Survey, but not later than 15-04-2000. The second routine annual survey was completed to the satisfaction of RS class surveyors and approval assigned on 19January1999. D 1.2.5 Occasional Hull Survey At the request of the owners, an additional hull survey was carried out by RS surveyors at St. Petersburg, Russia, in August 1999. This survey (designated occasional by RS) was to ascertain the vessel's compliance with requirements of the International Maritime Dangerous Goods Code and the Code of Safe Practice for Solid Bulk Cargoeswith respect to the possible future carriage of various types of ammonium nitrate fertilizers. At this time, the external survey found no newly incurred structural damage to the hull. The cargo holds and hatch covers were inspected and found in order. It was also confirmed that the vessel carried an updated Stability and Strength Booklet (operatingmanual)approved by RS on 20January1999. Approval for the carriage of ammonium nitrate fertilizers in cargo holds1and4 was assigned by RS surveyors on 09August1999. D 1.3 Hull Construction The hull was generally constructed with shipbuilding quality Grade A mild steel, while the main deck port and starboard side stringer plates and side shell sheer strakes were made of more notch (fracture) resistant GradeD steel. The layout of the principal longitudinal structural members in way of the cargo hatches, holds, upper wing ballast tanks, and double bottom tanks is shown in Figure11. TYPICAL CROSS SECTION PRINCIPAL CONSTRUCTION DETAILS Figure11. Typical cross section: principal construction details D 1.4 Post-Occurrence Hull Survey A close-up visual inspection of the damaged main deck and the internal structure of four of the upper wing water ballast tanks was carried out while the vessel was aground. Inspections of the side shell, side framing, inner bottom and transverse bulkheads in way of all of the cargo holds were carried out when the vessel was refloated and unloaded. The internal structure of the cargo holds, including the inner bottom tank top and hopper tank sides, was found to be generally free of any significant localized damage or distortion that might have been incurred prior to this occurrence. Side framing and top and bottom bracket connections throughout the cargo holds were free of any localized impact damage or significant wastage, and the welded connections to the side shell and hopper side plating were in good order. Localized minor corrosion pitting was found in the external painted surfaces of the port and starboard sides of the main deck plating throughout the mid-length of the vessel. Exposed deck plating inside the line of hatches was extensively corroded due to breakdown of the original paint protection. The internal structure of port and starboard upper wing water ballast tanks2 and 3showed extensive and active corrosion, particularly in way of the sloping tank bottom plating and transverse web frames. Large areas of these structural members were affected by and covered with hard and loose scale, large quantities of which had fallen loose and accumulated at the bottom of the tank. Many of the most recently renewed main deck longitudinals were comprised of relatively short lengths of steel flat bar, and the original members were extensively corroded. The ends of the longitudinals, exposed in way of the principal main deck failures, showed fillet-welded connections to the main deck plating with irregular throat sizes and leg lengths, grooving and undercutting of the deck plating, and a lack of penetration. In addition to those in way of the principal main deck failures, a further 19fractures were found in the main deck longitudinals. Some of these fractures were in way of butt joints, where the welding had incurred preferential corrosion; others were in way of heavily-corroded and locally-thinned metal of the original longitudinals. Ultrasonic thickness gauging measurements were made of the main deck, upper wing water ballast tank internal structures and the side shell plating, both forward, aft, and immediately adjacent to the principal transverse hull fractures. Average thickness of the main deck plating was found to be generally 8%to10% less than when the vessel was originally built, while that immediately in way of the principal failures was, on average, some 15%to18% less, with some localized heavier pitting. Average thickness of the most recently renewed deck longitudinals was94%of their original size, while the most inboard and original members were generally reduced to75%,with localized reductions of as much as50%. The average thickness of the sheer strakes and side shell plating were generally 95% of the originally fitted structure, while that of the side shell immediately in way of the principal failures was some76%.Localized thickness of the side shell plating near the bottom of the upper wing ballast tanks and adjacent to the principal hull failure was 65%of that originally fitted. The upper strakes of the sloped bottom plating of the upper wing water ballast tanks were generally 75%to90% of the original thicknesses; however, the lower strakes and the webs of the transverse framing were reduced to 55%to60%. The lower strakes of the sloping plating immediately adjacent to the principal transverse failures in upper wing water ballast tanks2(S) and3(P) were locally reduced to some 35%of their original thickness. A transverse brittle fracture, some two metres in length, was found in the main deck plating in way of port side upper wing water ballast tank3, immediately aft and parallel to the bulkhead at frame108. The fore and aft location of this fracture is coincident with the principal failure on the starboard side of the main deck. However, the propagation of the port side fracture was arrested where it reached the circular welded connection of a sounding pipe deck penetration fitting. Several small transverse fractures, which occurred in the port and starboard sides of the main deck plating prior to the principal hull failure, were located at that time; their progress was halted by drilling crack-arresting holes at their ends. Subsequent internal inspections showed that these fractures were generally in way of deck longitudinals and coincided with sudden discontinuities in their weld connections or thickness, and also in way of fractured or corroded butt welds. Close-up examination of the exposed ends of the principal fractures across the port and starboard sides of the main deck plating and in the side shell plating showed chevron-shaped markings indicative of rapid brittle fracture failure. The pattern of the markings indicated that the principal structural failures were initiated in the main deck plating in way of uneven, undercut or deteriorated fillet-welded connections of deck longitudinals and adjacent, localized corrosion pitting. The failure pattern also indicated that the fractures propagated from the deck plating into the deck longitudinals and down the side shell plating. The brittle fractures continued until the tensile loading induced by the hogging bending moment which acted on the hull, passed below the neutral axis of the midship section structural modulus and became compressive in nature. D 1.5 Sister Ships Subsequent to the Alcoroccurrence, the three sister ships (the Cheetah, the Lynx, and the AghiosNicolaos) were identified as ships of particular interest (SPI) by Transport Canada Marine Safety (TCMS). This in effect red flagged these vessels and targeted them for more detailed scrutiny if and when they entered Canadian waters. As an SPI, the Cheetahunderwent a PSC inspection in the Port of Sept-les, Quebec, on 07April2000. PSC officers found fractured deck longitudinals in various locations in way of top side water ballast tanks2and3. As well, numerous cracks on deck appeared to be emanating from the same fractured deck longitudinals beneath. Certain sections of the longitudinals had been recently renewed and fillet welds, joining the old with the new, were the origin of the initial cracks. Finally, six web frames in way of the top side tanks were found corroded and in need of insert plates. The Lynx was in Canadian waters within two weeks of the Alcor grounding and TSB investigators had an opportunity to board this vessel. The steering station was in all respects similar to that found on the Alcor, and the condition of the No2 top side water ballast tanks was markedly better. Tank coatings here were relatively intact and cathodic protection was in use. (The crew was in the process of changing the zinc anodes during the TSB visit.) The AghiosNicolaoshas not been reported within Canadian waters since the Alcorgrounding. Appendix A: Helm Simulations AppendixB: Canadian Transportation Agency Pilotage Review Report-1999 Section 157 of the Canada Marine Act, which came into force on 01October1998, contained a provision that amended the Pilotage Act by adding a requirement for the Minister to further review the pilotage system. The impetus for this review stemmed from the 1995National Marine Policy, which recognized a need for further analysis of some of the more contentious issues within the current pilotage regime. The objective of the review was to conduct a forward-looking examination of the marine pilotage system and to develop recommendations to ensure Canada has an efficient, viable, and safe pilotage system to meet the ongoing and long-term expectations and demands of all users. The parameters set forth in the legislation were fairly precise as to what issues were to be reviewed. Specifically, five distinct subject areas were covered: pilot certification process for masters and officers; training and licensing requirements for pilots; compulsory pilotage area designations; dispute resolution mechanisms; and measures taken in respect of financial self-sufficiency and cost reduction. On 11 August 1998, the Minister wrote to the Chair of the Canadian Transportation Agency (CTA), tasking the CTA with the conduct of the review. The CTA received written submissions, convened two national meetings, and held regional consultations with interested parties. All facets of the marine industry were active participants at many of these sessions. The CTA provided its final report to the Minister on 31August1999. The CTA review contains 21recommendations, all with which Transport Canada concurs in principle. Certain recommendations cannot be implemented exactly as submitted, and some of the proposed regulatory changes will require further legal scrutiny before being adopted... AppendixC: Outline General Arrangement Appendix D - Glossary 1. Units of measure in this report conform to IMO standards or, where there is no such standard, are expressed in the International System of units. 2. SeeAppendixD - Glossary for all abbreviations and acronyms. 3. All times are eastern standard time (Coordinated Universal Time minus five hours) unless otherwise noted. 4. All speeds are over the bottom unless otherwise noted. 5. When reconstructing the timeline, the time stamp of 1438:25 was used from the pilot's transmission on VHF radio when warning the downbound vessel of problems with the Alcor. Events were then laid down before and after this transmission in order to situate these events on a common timeline. All other times, with the exception of 1438:25, are approximate. 6. International Convention for the Safety of Life at Sea 1960, Regulation 29.The Alcor did not have to meet the time limits imposed by Regulation 29, because only passenger ships had this requirement. 7. Canadian Hydrographic Service / Fisheries and Oceans Canada, Atlas of Tidal Currents - Cap de Bon-Dsir to Trois-Rivires, 1997. 8. Canadian Hydrographic Service,Sailing Directions- St. Lawrence River ATL 112,1st ed., 1992. 9. Canadian Hydrographic Service / Fisheries and Oceans Canada, 1999 Canadian Tide and Current Tables, Volume 3, 1999. 10. All switches mentioned are two- or three-way rotary switches. 11. Detent torque is the actuating force required to turn the switch through its detented positions. 12. TSB Engineering Laboratory Report LP 122/99. 13. The Corporation des Pilotes du Bas Saint-Laurent represents the interests of the pilots in District2. The interests of the pilots in District1 are represented by the Corporation des Pilotes du Saint-Laurent Central (Central St.Lawrence Pilot Corporation). Contractual agreements entered into with the LPA by each pilot corporation are distinct. 14. Statutory Orders and Regulations (SOR) 83-274. 19. Pilot Corporation Board of Directors. 20. Computerized simulations were conducted on behalf of the TSB by the Fisheries and Marine Institute of Memorial University (Newfoundland). SeeAppendixA for further details. 21. Because of the electro-hydraulic arrangement, the wheel requires little effort to move and can be spun rapidly, if desired. 22. United States, National Transportation Safety Board, Report No RAR-98-01, Train Derailment; Australia, Marine Incident Investigation Unit, Report No119, Grounding of the Bulk Carrier Thebes, 11June1997. 23. American Bureau of Shipping, Guidance Notes on the Application of Ergonomics to Marine Systems, ABS, 1998, p. 16. 24. Captain R.W. Rowe, The Shiphandler's Guide, The Nautical Institute, 1996, p. 33. 25. Daniel H. MacElrevey, Ship-handling for the Mariner, 2nded., Cornell Maritime Press, Maryland, USA, 1988, p.8. 26. Daniel H. MacElrevey, Ship-handling for the Mariner, 2nd ed., Cornell Maritime Press, Maryland, USA, 1988, p.9. 27. Captain R.W. Rowe, The Shiphandler's Guide, The Nautical Institute, 1996, p.33. 28. Ship-handling Course Manual, Port Revel, Saint-Pierre-de-Bressieux, France, p.51. 29. IMO subcommittee on Standards of Training and Watchkeeping, STW 30/4, 24November1998 - Revision of Resolution A.485 (XII). 30. TSB Recommendation M99-06 (TSBReportM97W0197 [Raven Arrow]). 31. All depths are above chart datum unless otherwise specified and do not include tidal variations. 32. Clinker is listed in the Workplace Hazardous Materials Information Systems as a Class E (corrosive) material with a pH in water of 11 to 13. 33. The Minister of Fisheries and Oceans is empowered through the Oceans Act and a memorandum of understanding with TC to assume the powers as specified in the Navigable Waters Protection Act. 34. United States Coast GuardRisk-Based Decision-Making Guidelines, 2nd ed., vol. 3, chapter 1, 2001. 35. s.678 (1), Canada Shipping Act S.C., 1993, c. 36, s.16. 36. s.16, Navigation Waters Protection Act R.S., c.19, s. 14. 37. TSBReportNoM97L0030 (Venus). 38. TSBReport NosM97W0197 (Raven Arrow), M97W0022 (Hoegh Merit), and M99W0058(CapeAcacia). 39. Part C of the report deals in depth with the issue of closing and reopening of the channel. 40. During the first and second refloating efforts, situational awareness was further hampered by darkness and by restricted visibility, respectively. 41. Four pilots were present during the second refloating; two on the bridge of the Alcor and two on the lead tug. 42. The accuracy of position can range from 1m to 5m. 43. J. K. Pollard, E. D. Sussman, and M. Steams, Shipboard Crew Fatigue, Safety and Reduced Manning. Transportation Safety Center, United States Department of Transportation, DOT-MA-RD-84090014, 1990. 44. David F. Dinges, The Performance Effects of Fatigue, Proceedings of the Fatigue Symposium, National Transportation Safety Board and NASA Ames Research Center, 1995. 45. Mark R. Rosekind, Philippa L. Gander, Linda J. Connell, et al. Crew Factors in Flight Operations X: Alertness Management in Flight Operations Education Module. NASA Ames Research Center, 2001. 46. TSB Report Nos M93C0003(Nirja), M97W0197(RavenArrow), and M97L0030(Venus) 47. Great Lakes Pilotage Authority, Work Regulations and Assignment Procedures, Annex J-1. 48. Notices to Mariners, annual edition, April 2000. 49. Canada Shipping Act, s. 562.18. 50. Vessel Traffic Services Zones Regulations, s. 6(1)(c) and (f). 51. Section 2.5.2.1, Annex 1, International Maritime Organization Resolution A.857(20), 27 November 1997. 52. TSB Report Nos M90M4053, M92W1057, M92L3028, M95W0187, and M97N0071. 53. TSB Report No SM 95/01, A Safety Study of the Operational Relationship between Ship Masters/Watchkeeping Officers and Marine Pilots. 55. TSB Report Nos M90M4053, M92W1012, M92W1022, M92W1066, and M94W0064. 56. Great Britain, Marine Accident Investigation Branch, Safety Digest, vol.2, 1999. 57. C.J. Parker, Managing Risk In Shipping, The Nautical Institute, 1999, p. 12. 58. TSB Report NoM95L0147 (Dorado). 59. International Convention on Salvage, 1989, article 9. 60. TSB Report NosM97W0197(RavenArrow), M97L0030(Venus), M97C0120(OlympicMentor), and M93L0001(CanadianExplorer). 61. TSB recommendation M99-06, (TSB Report NoM97W0197 [Raven Arrow]). 62. Review of Pilotage Issues, August 1999. 63. Letter from the Minister of Transport to the Chairman of the Transportation Safety Board of Canada, dated 22January2000. 64. TP 13145E, prepared by KPMG Consulting for the Transportation Development Centre, Transport Canada, January 1998.